[
https://issues.apache.org/jira/browse/LUCENE-2089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12781179#action_12781179
]
Robert Muir commented on LUCENE-2089:
-------------------------------------
bq. Basically, what I'm saying is the old FuzzyQuery is just garbage really -
I've even seen an email were Doug talked about just dropping it.
I agree with garbage, but we cannot completely replace it anyway. for example
what if someone supplies a term of length 54 and asks for distance of 0.5?
we should not use this algorithm for nonsense like that, in that case I think
they should just use the garbage algorithm.
Here is a quote from that moman page:
It means that in theory, you could ask for a Levenshtein distance of 27! Well,
if you have a week ahead of you...
we shouldnt burn cycles creating useless tables that will be huge arrays either
in fuzzyquery, or whatever. we can't compute all the way up to infinity, this
is why i think something like 1,2,3 is "reasonable" , if it requires more edits
than that, go with the old brute force algorithm?
> explore using automaton for fuzzyquery
> --------------------------------------
>
> Key: LUCENE-2089
> URL: https://issues.apache.org/jira/browse/LUCENE-2089
> Project: Lucene - Java
> Issue Type: Wish
> Components: Search
> Reporter: Robert Muir
> Assignee: Mark Miller
> Priority: Minor
>
> Mark brought this up on LUCENE-1606 (i will assign this to him, I know he is
> itching to write that nasty algorithm)
> we can optimize fuzzyquery by using AutomatonTermEnum, here is my idea
> * up front, calculate the maximum required K edits needed to match the users
> supplied float threshold.
> * for at least common K (1,2,3, etc) we should use automatontermenum. if its
> outside of that, maybe use the existing slow logic. At high K, it will seek
> too much to be helpful anyway.
> i modified my wildcard benchmark to generate random fuzzy queries.
> * Pattern: 7N stands for NNNNNNN, etc.
> * AvgMS_DFA: this is the time spent creating the automaton (constructor)
> ||Pattern||Iter||AvgHits||AvgMS(old)||AvgMS (new,total)||AvgMS_DFA||
> |7N|10|64.0|4155.9|38.6|20.3|
> |14N|10|0.0|2511.6|46.0|37.9|
> |28N|10|0.0|2506.3|93.0|86.6|
> |56N|10|0.0|2524.5|304.4|298.5|
> as you can see, this prototype is no good yet, because it creates the DFA in
> a slow way. right now it creates an NFA, and all this wasted time is in
> NFA->DFA conversion.
> So, for a very long string, it just gets worse and worse. This has nothing to
> do with lucene, and here you can see, the TermEnum is fast (AvgMS -
> AvgMS_DFA), there is no problem there.
> instead we should just build a DFA to begin with, maybe with this paper:
> http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.16.652
> we can precompute the tables with that algorithm up to some reasonable K, and
> then I think we are ok.
> the paper references using http://portal.acm.org/citation.cfm?id=135907 for
> linear minimization, if someone wants to implement this they should not worry
> about minimization.
> in fact, we need to at some point determine if AutomatonQuery should even
> minimize FSM's at all, or if it is simply enough for them to be deterministic
> with no transitions to dead states. (The only code that actually assumes
> minimal DFA is the "Dumb" vs "Smart" heuristic and this can be rewritten as a
> summation easily). we need to benchmark really complex DFAs (i.e. write a
> regex benchmark) to figure out if minimization is even helping right now.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: java-dev-h...@lucene.apache.org
